Classifications

• • H—ELECTRICITY
  • H03—ELECTRONIC CIRCUITRY
  • H03F—AMPLIFIERS
  • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
  • H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
  • H03F3/21—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only
  • H03F3/217—Class D power amplifiers; Switching amplifiers
  • H03F3/2178—Class D power amplifiers; Switching amplifiers using more than one switch or switching amplifier in parallel or in series
• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
  • H02M3/00—Conversion of dc power input into dc power output
  • H02M3/02—Conversion of dc power input into dc power output without intermediate conversion into ac
  • H02M3/04—Conversion of dc power input into dc power output without intermediate conversion into ac by static converters
  • H02M3/10—Conversion of dc power input into dc power output without intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode
  • H02M3/145—Conversion of dc power input into dc power output without intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal
  • H02M3/155—Conversion of dc power input into dc power output without intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only
  • H02M3/156—Conversion of dc power input into dc power output without intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only with automatic control of output voltage or current, e.g. switching regulators
  • H02M3/158—Conversion of dc power input into dc power output without intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only with automatic control of output voltage or current, e.g. switching regulators including plural semiconductor devices as final control devices for a single load
  • H02M3/1582—Buck-boost converters
• • H—ELECTRICITY
  • H03—ELECTRONIC CIRCUITRY
  • H03F—AMPLIFIERS
  • H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
  • H03F1/02—Modifications of amplifiers to raise the efficiency, e.g. gliding Class A stages, use of an auxiliary oscillation
  • H03F1/0205—Modifications of amplifiers to raise the efficiency, e.g. gliding Class A stages, use of an auxiliary oscillation in transistor amplifiers
  • H03F1/0211—Modifications of amplifiers to raise the efficiency, e.g. gliding Class A stages, use of an auxiliary oscillation in transistor amplifiers with control of the supply voltage or current
  • H03F1/0216—Continuous control
  • H03F1/0233—Continuous control by using a signal derived from the output signal, e.g. bootstrapping the voltage supply
  • H03F1/0238—Continuous control by using a signal derived from the output signal, e.g. bootstrapping the voltage supply using supply converters
• • H—ELECTRICITY
  • H03—ELECTRONIC CIRCUITRY
  • H03F—AMPLIFIERS
  • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
  • H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
  • H03F3/21—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only
  • H03F3/217—Class D power amplifiers; Switching amplifiers
  • H03F3/2171—Class D power amplifiers; Switching amplifiers with field-effect devices
• • H—ELECTRICITY
  • H03—ELECTRONIC CIRCUITRY
  • H03F—AMPLIFIERS
  • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
  • H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
  • H03F3/21—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only
  • H03F3/217—Class D power amplifiers; Switching amplifiers
  • H03F3/2173—Class D power amplifiers; Switching amplifiers of the bridge type

Definitions

• the invention relates to switching amplifiers and switching power converters with bipolar outputs.
• the invention relates to amplifier circuits that are able to provide for power efficiency, space efficiency and cost efficiency over known amplifier circuits.
• FIG. 1 shows the single-ended connection of a load 10 to a unidirectional amplifier 12 in accordance with the prior art. As in many applications, only a single power supply rail is available. The output signal seen by the load is made DC-free by including a DC-blocking capacitor 14. This is shown connected to zero volts, but can alternatively be connected to the amplifier's other power-supply terminal, VP.
• the amplifier 12 has been drawn as a power converter to emphasize the power flow, which is always from the power supply to the load. The input signal and the control circuitry are left un-shown. The amplifier 12 may for example have a class-B output stage (non-switching).
• FIG. 1 shows the single-ended connection of a load 10 to a unidirectional amplifier 12 in accordance with the prior art. As in many applications, only a single power supply rail is available. The output signal seen by the load is made DC-free by including a DC-blocking capacitor 14. This is shown connected to zero volts, but can alternatively be connected to the amplifier'
• the differential output voltage range lies between -VP (minus VP) and +VP (plus VP). For example, with VP equal to 10 volts, the range lies between -10 volts and +10 volts.
• the circuit shown in FIG. 2 cannot boost the output beyond -VP and +VP.
• FIG. 3 shows a single-ended switching amplifier that acts as a bidirectional power converter 32.
• the bidirectional power converter 32 is connected to a load 30.
• a DC- blocking capacitor 34 is also included.
• the power converter 32 is typically a buck converter.
• Such amplifiers can drive loads that are partly or wholly reactive, without consuming a lot of power. Reactive energy from the load, or from the amplifier's output filter, is passed back to the power supply. This energy does then need to be accommodated, or there may be excessive supply voltage pumping, but there are well established solutions for that.
• FIG. 4 shows a bridged switching amplifier that uses two bidirectional power converter blocks 42, 44 bridging a load 40. If the power converters are buck converters, this arrangement becomes the classic class-D H-bridge. It is shown, from a power-conversion viewpoint, in figures 6.11 and 6.14.5 of Erickson & Maksimovic. As before, the input signal and the control circuitry are left un-shown.
• the input signal can take many forms, including analog waveforms and digital number streams.
• the control scheme can take many forms, including pure feedforward, single-loop voltage feedback, single-loop current feedback, charge control, multi-loop, and hybrid.
• US6646507 discloses a 'Power booster amplifier' which intrinsically achieves voltage boosting of up to x2.
• Hwu and Yau present a circuit with voltage boosting of up to x3.
• the circuits are intricate, and the boosted voltages are present at all times. This gives reduced efficiency during those periods in which voltage boosting is not needed, which is typically most of the time.
• class HD sometimes called class HD.
• Selective use of boosted voltages can very much improve the energy efficiencies that are achieved at low signal levels. In applications where the signals tend to have high crest factors, this benefit tends to map directly to increased battery life.
• class HD circuits generally have at least two largely separate conversion blocks, each with its own wound component. This makes for relatively high cost and size, and low efficiency.
• An ideal solution would be single-stage and comprise fewer components.
• the power conversion literature describes a different way of raising efficiencies at low signal levels, which is to design the converter to use DCM (discontinuous conduction mode).
• DCM discontinuous conduction mode
• each cycle of switching activity is separated from the next by a period of time in which the main switch currents are zero.
• CCM continuous conduction mode
• DCM discontinuous conduction mode
• boost-capable blocks are Cuk converters (figure 6.15.1 in Erickson & Maksimovic). Cuk and Erickson presented a related paper in 1978 titled ⁇ conceptually new high-frequency switched-mode power amplifier technique eliminates current ripple' (Powercon5 conference proceedings). In it they disclose the possibility of instead using boost converters or inverting buck-boost converters (figures 6.14.2 and 6.14.3 in Erickson & Maksimovic). Caceres and Barbi also present the use of boost converters in their 1995 IEEE paper, ⁇ boost DC-AC converter: Operation, analysis, control and experimentation' (IEEE IECON conference proceedings). Such solutions are appealing, but they still comprise two largely- separate blocks, each with its own wound component. They are not single-stage designs.
• a switching amplifier needs to be bidirectional. More particularly, it needs to be a four-quadrant amplifier, managing seamlessly as the polarities of the load current and load voltage move between (+ +), (- +), (- -) and (+ -). In the (+ +) and (- -) quadrants the amplifier delivers energy to the load. In the (+ -) and (- +) quadrants it recycles energy back from the load.
• Erickson & Maksimovic does not explicitly teach about four-quadrant conversion, but the topic is covered in chapter 9 of the 201 1 book 'Energy Storage in the Emerging Era of Smart Grids', edited by Carbone (ISBN 978 953 307 269 2).
• the diagonal load line 50 is for purely resistive loads.
• the circular load line 52 is for sine waves into purely-reactive loads. Most loads are neither purely-resistive nor purely reactive, especially when seen through an output filter. The recycling of reactive energy from output filters can be just as important as the recycling of reactive energy from loads.
• DCM DCM
• a control block selects a switching sequence from a small number of available sequences, once every control cycle. The selection depends in-part on the present state of the circuit. For example, it can depend on the polarities of the load voltage and the load current. If the set of available switching sequences has got blind spots, or if the control block selects inappropriate switching sequences from the set, the circuit will not function as a four- quadrant amplifier.
• a troublesome feature of the circuits disclosed in US5610553 and US8410851 is that some of the output voltages necessarily go below zero volts. This makes it difficult or impossible to implement these circuits as semiconductor ICs (integrated circuits). In most IC processes, it is not allowed for any node to go more negative than the substrate.
• An attractive feature of the circuits disclosed in US5610553 and US8410851 is that they use just one wound component; a simple two-terminal inductor. The same is true of various circuits from the DC DC converter world. Some examples can be found in the 2001 paper on 'Single-inductor multiple-output switching converters' by Ki and Ma (IEEE Transactions on Circuits and Systems II). A further example features in the 2011 paper titled 'Proposal of a low power, 1.6 MHz, 91 % efficiency, single inductor, double
• a switching amplifier or bipolar-output switching power converter comprising: a power stage comprising a first terminal, a second terminal and a common terminal; wherein the power stage is configured to operate with a load circuit connected between the first and second terminals.
• the amplifier or power converter is configured to span the load circuit. This configuration provides for improvements in power efficiency, space efficiency and cost efficiency over the prior art. The improvements can be obtained across a wide range of signal amplitudes, and with load circuits that may be resistive, partially reactive or wholly reactive. As a consequence of the spanning configuration, the amplifier or power converter can establish both positive and negative differential voltages at its load.
• the power stage provides greater than x2 voltage boosting and signal modulation in a single stage.
• the overall amplifier then combines voltage boosting and signal modulation in a single stage. This avoids the inefficiencies and costs of solutions that do all of their voltage boosting in a first block, and then modulate the signal onto the boosted voltage in a second block.
• the power stage is configured, at some times, to move energy from the first terminal to the second terminal, and at other times, to from the second terminal to the first terminal. Energy can be recovered from the load circuit, providing for power efficiency.
• the amplifier is configured to operate in discontinuous conduction mode (DCM). DCM provides high overall efficiency when amplifying signals with high peak-to- average power ratios. DCM also removes dead-time distortion effects that are a problem with conventional class-D amplifiers.
• the power stage may comprise a plurality of switches that can be operated to provide both voltage boosting of a voltage from a power supply and signal modulation based on an input signal.
• the power stage is advantageously configured to connect to a power supply providing a power supply voltage such that neither the first terminal nor the second terminal is permanently tied to the power supply voltage.
• the first terminal is connected to a first node and the second terminal is connected to a second node, wherein the first node and the second node may be dynamically tied to the power supply voltage.
• the first node is tied to the power supply voltage
• the second node is tied to the power supply voltage or neither the first node nor the second node is tied to the power supply voltage.
• a first diode may be connected from the power supply voltage to the first node and a second diode may be connected from the power supply voltage to the second node.
• the first and second diodes may allow for a flow of current from the power supply to either the first node or the second node, depending on a voltage at the first and second nodes.
• the amplifier may further comprise a first reservoir capacitor connected to the first node and a second reservoir capacitor connected to the second node.
• the amplifier operates to move energy from the reservoir capacitors to the load and from the load to the reservoir capacitors.
• the first and second reservoir capacitors may be connected to ground.
• a first transistor may be connected to the first node and a second transistor connected to the second node, the first and second transistors allowing for a flow of current between the power supply and either the first node or the second node (or neither node).
• reservoir capacitors may not be required as energy can be moved from the load back to the power supply.
• the power supply then acts partly as an energy reservoir.
• the power stage is bidirectional in that energy can be moved from the first terminal to the second terminal and from the second terminal to the first terminal.
• the power stage comprises a bidirectional buck-boost converter.
• the power stage may take other forms that provide both voltage boosting and voltage bucking in both directions.
• the bidirectional power stage may comprise two unidirectional power converters arranged to provide bidirectional power conversion. For example, an arrangement of two interleaved, unidirectional buck-boost circuits may form the power stage.
• the two unidirectional power converters may be cross-coupled.
• the two unidirectional power converters may share a common inductor.
• the bidirectional buck-boost converter may take a plurality of different forms.
• the bidirectional buck-boost converter is a bidirectional flyback converter.
• the bidirectional flyback converter may comprise a coupled inductor and two active switches. The small number of active switches means that losses are relatively small.
• the active switches may be field effect transistors (FETs).
• FETs field effect transistors
• the FETs may be ground- referred, which makes them easy to drive.
• the bidirectional buck-boost converter may comprise a plurality of active switches.
• the bidirectional buck-boost converter comprises no more than four active switches.
• the active switches may comprise transistors. Schottky diodes may be connected across each of the transistors.
• the bidirectional buck-boost converter may be a bidirectional four-switch converter.
• the bidirectional four-switch converter may advantageously comprise a single, two-terminal inductor.
• the bidirectional buck-boost converter may be a three-port converter comprising a further terminal, wherein the further terminal is connected to the power supply voltage.
• a three port power converter may deliver reduced losses in some circumstances.
• the bidirectional buck-boost converter comprises a single wound component.
• the bidirectional buck-boost converter is a non-inverting converter.
• the bidirectional buck-boost converter is configured for four-quadrant operation.
• the bidirectional buck-boost converter is configured to provide voltage boosting to greater than twice a supply voltage. This is advantageous in audio applications, with high peak-to- average power ratios.
• the amplifier may be provided in an integrated circuit. The amplifier may be controlled so that the voltage at the first terminal and the voltage at the second terminal never drop below a substrate voltage, which is typically zero volts.
• the amplifier may further comprise a control circuit for the power stage.
• the control circuit may be configured to control the power stage based on a voltage across the load circuit or current through the load circuit.
• the control circuit may operate to control a differential voltage across the load circuit, or may operate to control another parameter, such as a current, an integrated voltage, charge, or some combination of these parameters.
• the bidirectional buck-boost converter may comprise a plurality of active switches and the control circuit may be configured to control each of the active switches.
• a switching amplifier circuit or bipolar-output switching power converter circuit comprising: an amplifier according to the first aspect of the invention; and a load circuit connected between the first and second terminals.
• the load circuit typically comprises an output filter and a load.
• the load circuit may comprise at least one reactive component, and the control circuit may be configured to operate the power stage to take energy from the load circuit and to subsequently reapply that energy to the load circuit.
• the load circuit may take a plurality of different forms.
• the load circuit may comprise a load alone or a load together with additional components, such as an output filter.
• the load may comprise, for example, a piezoelectric actuator or a loudspeaker.
• the load circuit may comprise a predominantly resistive load and a capacitive output filter. It may additionally comprise an inductive component, for low-pass filtering.
• the amplifier circuit may further comprise a power adaptation block and a power reservoir connected to the power supply voltage. This may be used to smooth energy demand from a primary power supply and ease power demand bottlenecks.
• the amplifier may be configured such that the first and second terminals are at positive voltage throughout operation. This allows the amplifier circuit to be implemented in an integrated circuit, on a semiconductor chip.
• a method of operating a switching amplifier or bipolar-output switching power converter comprising a power stage having a first terminal, a second terminal and a common terminal, the method comprising: connecting a load circuit between the first terminal and the second terminal.
• the method may comprise the step of operating the power stage to move energy from the first terminal to the second terminal, and from the second terminal to the first terminal.
• the power stage comprises a bidirectional buck-boost converter.
• the method may comprise controlling the bidirectional buck-boost converter based on a voltage or current at the first terminal and a voltage or current at the second terminal.
• the method may comprise operating the bidirectional buck-boost converter in discontinuous conduction mode.
• the bidirectional buck-boost converter may be connected to a power supply voltage such that each of the first terminal and the second terminal can be dynamically connected to and disconnected from the power supply voltage.
• the load circuit may comprise a reactive component, and the method may further comprise taking energy from the load circuit and subsequently reapplying that energy to the load circuit.
• a switching amplifier circuit comprising: a bidirectional power stage; a load circuit connected to the bidirectional power stage; a power supply supplying a power supply voltage to the bidirectional power stage; and at least one energy storage device connected to the bidirectional power stage but not directly connected to the power supply, wherein, in operation, the bidirectional power stage can be controlled to transfer energy from the load to the energy storage device and from the energy storage device to the load.
• the bidirectional power stage may comprise a buck-boost converter and is preferably non- inverting.
• the load circuit is advantageously fully or partially reactive. The load circuit or the load may be connected between the two terminals of the bidirectional power stage.
• the terminals of the bidirectional power stage are each dynamically connected to the power supply voltage.
• the amplifier or power converter is preferably configured to provide voltage boosting and signal modulation in a single stage.
• FIG. 1 depicts a prior-art single-ended amplifier or power converter with DC blocking
• FIG. 2 depicts a prior-art bridged amplifier or power converter
• FIG. 3 depicts a bidirectional prior-art single-ended amplifier or power converter
• FIG. 4 depicts a bidirectional prior-art bridged amplifier or power converter
• FIG. 5 shows the four quadrants of amplifier operation, plus two load lines
• FIG. 6 depicts the achievable operating area of one particular prior-art amplifier
• FIG. 7 depicts a prior-art amplifier with bipolar output and grounded load
• FIG. 8 depicts a prior-art amplifier with bipolar output and floating load
• FIG. 9 shows a switching amplifier or power converter according to the invention.
• FIG. 10 shows the circuit of FIG. 9 incorporating a bidirectional buck-boost converter
• FIG. 1 1 shows a diode-based power injection arrangement for the circuit of FIG. 10
• FIG. 12 shows a FET-based dynamic power supply voltage pinning arrangement for the circuit of FIG. 10;
• FIG. 13 shows a bidirectional flyback converter in the circuit of FIG. 12
• FIG. 14 shows a bidirectional four-switch converter in the circuit of FIG. 12;
• FIG. 15 shows some example load circuits;
• FIG. 16 illustrates an input signal and control block of an amplifier in accordance with the invention
• FIG. 17 shows a selection of bidirectional buck-boost converters
• FIG. 18 shows the circuit of FIG. 10 comprising a three-port converter
• FIG. 19 shows a selection of three-port converters that may be used in the circuit of FIG. 18;
• FIG. 20 shows an alternative converter that can be used in the circuit of FIG. 9; and FIG. 21 illustrates an optional power adaption block and power reservoir.
• FIG. 9 The key connections of an amplifier or power converter according to the present invention are shown in general form in FIG. 9.
• a power stage 92 is connected to a load circuit 90 comprising a load, via a first node (N1) 94 and a second node (N2) 96.
• the power stage 92 is also connected to a power supply (not shown) via two or more power supply nodes 97, 98, which in this case have voltages VP and 0V (zero volts).
• Nodes 94 and 96 are distinct from the power supply nodes 97, 98 in FIG.
• the amplifier establishes a desired signal at the load circuit 90 or at the load by, at some times transferring energy from N1 to N2, and at other times transferring energy from N2 to N1.
• the power stage's voltage conversion range includes both step-down (buck) and step-up (boost).
• the amplifier or power converter can establish both positive and negative differential voltages at the load. It can provide a bipolar output, even though its power stage only has unipolar input/output (IO) at N 1 and N2.
• IO input/output
• the amplifier or power converter can be operated as a four-quadrant device. When operated this way, it can recover inductive energy from the load circuit 90, and it can successfully drive highly reactive loads.
• the voltage step-up capability of the power stage 92 preferably extends considerably beyond voltage doubling. This gives the amplifier or power converter an overall voltage- boosting capability. It can establish differential voltages on the load that extend
• Voltage-boosting amplifiers and power converters according to the configuration of FIG. 9 combine their voltage boosting and signal modulation in a single power stage. They avoid the inefficiencies and costs of solutions that do all of their voltage boosting in a first block, and then modulate the signal onto the boosted voltage in a second block.
• the energy path between N1 and N2 is preferably a direct path, meaning that it involves just one stage or conversion rather than a cascade of multiple stages or conversions.
• the energy transfers are preferably direct transfers, preferably involving just one wound component.
• the power supply may be a single-rail DC supply (unipolar), or it can instead be multi-rail or AC (bipolar).
• the power supply and the power stage 92 have node 0V as a common node in FIG. 9, though they can instead be electrically isolated.
• VP may be a static voltage, or it may instead be time-varying or parameter-dependent. In general, power can flow not just from the power supply, but also back to it. As shown in FIG.
• bidirectional buck-boost converter is used herein to mean a bidirectional converter whose conversion range includes both voltage step-down and voltage step-up, in both directions of energy transfer.
• Bidirectional buck-boost converters are well-known in the art. However, they are generally used as two-port devices with their power sources and destinations connected to one port or the other port. This can be seen in US7786709, for example.
• the load is instead connected across the bidirectional buck-boost converter.
• Bidirectional buck-boost converters typically have unipolar input/outputs (lOs). They are sometimes called bidirectional buck-boost DC-DC converters.
• the term DC-DC relates to the unipolarity of the lOs. It does not mean that the IO voltages have to be static.
• Bidirectional buck-boost converters can generally move seamlessly from voltage step- down, through the 1 :1 ratio, to voltage step-up.
• the 1 : 1 ratio is somewhere in the middle of their voltage conversion range, not at one of the ends.
• the 1 : 1 ratio corresponds to zero differential voltage at the load.
• FIG. 1 1 and FIG. 12 show two of them.
• the heart of the power stage 92 is a bidirectional buck-boost converter 100, as already described with reference to FIG. 10. Its two IO terminals are connected to nodes N1 and N2 respectively, and its common terminal is connected to 0V.
• Four-terminal isolated bidirectional buck-boost converters can also be used, by first commoning a terminal from one side with a terminal from the other side. This action un-isolates the converter and creates a common terminal.
• FIG. 1 1 the heart of the power stage 92 is a bidirectional buck-boost converter 100, as already described with reference to FIG. 10. Its two IO terminals are connected to nodes N1 and N2 respectively, and its common terminal is connected to 0V.
• Four-terminal isolated bidirectional buck-boost converters can also be used, by first commoning a terminal from one side with a terminal from the other side. This action un-isolates the converter and creates
• diodes 1 14, 116 via which power can flow from the power supply VP. They are typically Schottky diodes. Diode 1 14 is connected from VP to N1 , and diode 1 16 is connected from VP to N2.
• the load circuit 90 is a simple two-terminal capacitive load 115.
• the arrangement also needs a first reservoir capacitor 118 between N1 and 0V (or VP), and a second reservoir capacitor 119 between N2 and 0V (or VP).
• the differential voltage at the load is (VN2-VN1).
• the bidirectional buck- boost converter 100 moves charge from N1 to N2.
• the bidirectional buck-boost converter 100 moves charge from N2 to N1.
• the circuit could output a sine wave without ever needing any top-up charge from the power supply.
• the circuit is not lossless. With a sine wave, the circuit takes top-up charge from the power supply through first diode 1 14 at-and-near positive peaks, and through second diode 116 at-and-near negative peaks. Representative voltage waveforms of VN1 and VN2 over time are shown in the windows 1 11 112 in FIG. 1 1. The difference between these voltages is a desired sine wave at the load.
• the bidirectional buck-boost converter 100 drives the waveforms by moving energy from one side of the circuit to the other.
• the first and second capacitors 1 18 and 119 act as reciprocating energy reservoirs.
• the energy transfers from the power supply to the circuit are not particularly critical.
• the transfers through the bidirectional buck-boost converter 100 i.e. between N1 and N2, are more significant. Energy makes the former journey no more than once, but for efficient operation it must make the latter journey many times.
• FIG. 11 has been described with reference to capacitive loads, it is also usable with other types of load. This is described below.
• FIG. 12 the heart of the power stage 92 is again a bidirectional buck-boost converter 100.
• the power stage is like the one in FIG. 1 1 , except that the two diodes have been replaced with two active switches 124, 126. Each of these is shown as a single FET, complete with body diode. Double-FET implementations to extend the voltage ranges at N1 and N2 further below VP are also possible.
• the first FET 124 is connected from VP to N1
• the second FET 126 is connected from VP to N2.
• Their orientation is such that the body diode does not conduct during normal operation.
• a preferred way of operating these FETs is as follows.
• FET 124 When the differential voltage at the load circuit (VN2 VN1) is at-least one diode-drop more positive than zero, FET 124 is held on. When it is at least one diode-drop more negative than zero, FET 126 is held on. When it is near or at zero, one of the FETs 124, 126 is held on, but which one may depend on signal history and dynamics. Implementing some hysteretic behaviour in this region can improve efficiency and waveform fidelity, especially at low signal levels.
• FETs 124 and 126 one side and the other side of the load circuit can be alternately pinned to the power supply voltage VP.
• Representative voltage waveforms of VN1 and VN2 over time are shown in windows 121 and 122 of FIG. 12. The difference between these voltages is a desired sine wave at the load.
• These FETs can be switched relatively infrequently. Their switching rate is signal-related, and approaches zero at low signal levels. It is much lower than the switching rate of the switches within the bidirectional buck- boost converter 100. This very-much eases the compromise between low on-resistance, low capacitance and low price, for these two FETs.
• the arrangement of FIG. 12 allows reactive power to flow back to the power supply. With AC signals and partly or wholly reactive load circuits, the power supply acts in-part as a reciprocating energy reservoir. The reciprocating energy and/or the associated supply- voltage pumping need to be accommodated.
• Bidirectional buck-boost converters are well-known in the art. Many different types exist that could be used in the circuits of FIGS. 10 to 12. FIG. 13 and FIG. 14 show two preferred types, applied to the exemplary arrangement of FIG. 12.
• the circuit of FIG. 13 is a first preferred embodiment of the present invention.
• the bidirectional buck-boost converter 100 is a basic bidirectional flyback converter.
• the implementation and operation of bidirectional flyback converters has been detailed in many publications. This includes for example a 1989 paper on the 'Current mode controlled bidirectional flyback converter' by Venkatesan (IEEE PESC convention proceedings).
• the bidirectional flyback converter is composed of a coupled inductor 130, 131 and just two active switches 132, 134. This is a relatively small number of active switches, which keeps losses down.
• the active switches 132, 134 are single FETs. They are ground referred, which makes them easy to drive.
• Schottky diodes 136 and 138 are shown connected across the FETs 132 and 134 respectively. As is well-known in the art, fitting these can be appropriate for any switching circuit that is not always operated in continuous conduction mode (CCM) with synchronous rectification. They perform better than the FETs' body diodes, which raises the obtainable efficiencies. Schottky diodes are not shown in FIG. 14 or any of the subsequent figures, partly because they are not always necessary, but mostly to improve the clarity of the figures. However, they may be used wherever a FET is shown.
• CCM continuous conduction mode
• FET 134 To move the energy from the load 90 back to the power supply FET 134 is first switched on. The current in FET 134 builds up as the inductor is energised by the load, through node N2. At an appropriate instant FET 134 is switched off. The current moves across to diode 136, and the energy starts to move from the inductor 130 out to the power supply, through node N1. The current winds down to zero, and then stays at zero until the next cycle.
• FIG. 14 shows a second preferred embodiment of the present invention.
• the bidirectional buck-boost converter 100 is an archetypal bidirectional four-switch converter.
• the four-switch converter has been referred to by a number of different names over the years. In Erickson & Maksimovic it is called the non-inverting buck-boost converter.
• bidirectional four-switch converters have been detailed in many publications. This includes for example a 1995 paper on '20 kW water-cooled prototype of a buck-boost bidirectional DC-DC converter topology for electrical vehicle motor drives' by Caricchi, Crescimbini and Napoli (IEEE APEC conference proceedings). Further teaching can be found in United States patent US5734258.
• the wound component in the bidirectional four-switch converter is a simple two-terminal inductor 140. This is an advantage over the bidirectional flyback converter.
• the number of active switches in the bidirectional four-switch converter is four. This is a disadvantage over the bidirectional flyback converter. In FIG. 14 the active switches are shown as single FETs, 142, 144, 146, 148. There will typically also be four corresponding Schottky diodes (not shown).
• One way of operating the bidirectional four-switch converter of FIG. 14 in DCM is as follows. To make the output voltage (VN2-VN1) more positive (or equivalently, less negative), FET 142 and FET 148 are switched on then off. While they are on, the current in them builds up as the inductor 140 is energised. When they are switched off, the current moves to the body diodes of FET 146 and FET 144, or to their associated Schottky diodes (not shown). It winds down to zero, and then stays at zero until the next cycle. To make the output voltage (VN2-VN1) more negative (or equivalently, less positive), FET 144 and FET 146 are switched on then off.
• This switch sequence provides the function of two uni-directional boost converters who share a single inductor.
• the two boost converters are cross coupled in that the input of one converter is connected to the output of the other converter.
• Bi-directional conversion, of Figure 9 is therefore achieved through the implementation of two uni-directional converters connected in opposing direction.
• Figure 20 illustrates a similar configuration though not sharing a common inductor. In all cases, when the voltage of either VN1 or VN2 falls below the battery voltage, Vp, the diode/switch 124 or 126 conducts to replenish lost charge.
• the load circuits 90 in FIGS. 9 to 14 may include connections to 0V, or to other circuit nodes. They typically includean output filter and either a load or terminals to connect a load.
• the output filter is typically comprised of a small number of passive components.
• FIG. 15 illustrates some examples.
• FIG. 15A shows the case of no output filter and a two-terminal capacitive load 150 (e.g. a piezoelectric actuator or an electroactive polymer actuator).
• FIG. 15B adds a VHF filter to reduce radio-frequency emissions.
• the filter comprises a common-mode choke 152, 153 and ceramic capacitors 154, 155 and 156, arranged in such a way that they do not affect the switching and baseband performance of the amplifier or power converter.
• FIG. 15C shows a resistive load 160 with a shunt capacitor 162 for smoothing.
• the shunt capacitor 162 makes the load circuit look capacitive at the switching frequency. This gives good results with the power stages of FIG. 13 and FIG 14.
• FIG. 15D shows a hybrid approach.
• the shunt capacitor 162 and the two 3-terminal capacitors 164, 165 for VHF suppression there is a second-order output filter comprising the load 160 plus inductor 166 and capacitor 168.
• FIG. 15E shows a three-terminal capacitive load 170 with its common terminal connected to 0V.
• Four-terminal loads are also possible, and can be accommodated for- example by commoning two of their terminals, so that they become three-terminal loads.
• the load circuit of FIG. 15E also comprises two ferrite beads 172, 173, for radio frequency suppression.
• FIG. 15E With a three-terminal capacitive load as in FIG. 15E, the arrangement of FIG. 1 1 would not need reservoir capacitors 118 and 119. The job of capacitors 118 and 119 would be done by the two halves of the load.
• FIG. 16 shows the load circuit 90 and power stage 92 of FIGS. 9 to 14, but it additionally shows the amplifier or power converter's control block 180, in general form.
• This block receives an input signal 181 which in some prescribed way defines a desired signal at the load. It generally also receives status signals and value information 182 from the power stage 92 and the load circuit 90. And it provides control signals 183 to drive the power stage's active switches.
• the control block implements a control scheme. Many control schemes are known in the art.
• a preferred control scheme for the circuits of FIG. 13 and FIG. 14, when driving a resistive load with dominant shunt capacitance as in FIG. 15C for example, is the energy-centric control scheme of Janocha and Stiebel.
• the control circuit of a power conversion block with a common terminal (0V) is generally configured to control a parameter associated with an aggregate power sink that is connected between another of that block's terminals and the common terminal.
• the controlled parameter would typically be the voltage at N2.
• the control circuit must be configured to establish the desired signal at the load circuit or at the load. As an example, it may be configured to control the differential voltage at the load circuit (VN2-VN1). More generally, the controlled parameter could be voltage, current, integrated voltage, charge, some other important measure, or some combination of these things.
• FIG. 17A shows a bidirectional flyback converter with P-type active clamps. It is from a 2010 paper titled ⁇ high efficiency flyback converter with new active clamp technique' by Zhang, Huang, Wu and Qian (IEEE Transactions on Power Electronics).
• FIG. 17C is a generalized representation of bidirectional flyback converters. The shaded area covers a plurality of different possibilities. It encompasses FIG.
• FIG. 17B shows a bidirectional low-stress buck-boost converter. It is from a 2001 paper on 'Buck-boost PWM converters having two independently controlled switches' by Chen, Maksimovic and Erickson (IEEE PESC conference proceedings). It can be derived by the well-known step of replacing the diodes of a unidirectional converter with active switches, or alternatively by combining figures 9 and 10 of that paper.
• FIG. 17D shows a bidirectional low-stress buck-boost converter. It is from a 2001 paper on 'Buck-boost PWM converters having two independently controlled switches' by Chen, Maksimovic and Erickson (IEEE PESC conference proceedings). It can be derived by the well-known step of replacing the diodes of a unidirectional converter with active switches, or alternatively by combining figures 9 and 10 of that paper.
• FIG. 17D shows a bidirectional low-stress buck-boost converter. It is from a 2001 paper on 'Buck-boos
• FIG. 17E shows the bidirectional noninverting Cuk converter. It can be found for-example in figure 22a of chapter 9 of the Carbone book (ISBN 978 953 307 269 2).
• FIG. 17F shows the split-pi converter, which was disclosed in GB2376357.
• FIG. 17G shows an advantageous bidirectional four-winding converter from figure 4 of chapter 8 of the Carbone book.
• FIG. 17H shows the bidirectional SEPIC or ZETA converter, which can be found for-example in figure 6.15 of Erickson & Maksimovic.
• FIG. 17 shows an amplifier or power converter according to the present invention, in which the power stage comprises a three-port converter 200.
• the three-port converter is connected to 0V, N1 and N2 as before, but it is also connected to VP.
• the three-port converter 200 has got all of the functionality of the bidirectional buck-boost converters described in relation to FIGS. 10 to 12. Additionally, it can transfer energy between VP and N 1 , and between VP and N2. Many designs of such three-port converters are possible.
• FIG. 19 shows four examples.
• FIG. 19A is similar to FIG. 17A.
• FIG. 19A additionally includes two abstracted switches 202, 204. By appropriately controlling these switches and the FETs, the control block can transfer energy from and to VP.
• the circuit of FIG. 19B achieves the same result using fewer switches, but it requires a more complex wound component 200.
• FIG. 19C and FIG. 19D the path from VP is unidirectional and self-regulating.
• the converters of FIGS. 19A to 19D demand the presence of reservoir capacitors like those shown in FIG. 1 1. Another possibility is to add further windings to the wound components in FIG. 17, and to use those as a path from or back to the power supply.
• FIG. 20 shows another circuit that can be used in the present invention. Unlike the circuits in FIG. 17, it is not intrinsically a bidirectional buck-boost converter. It is an interleave of two unidirectional buck-boost circuits. The FETs and diodes connected to the upper coil can transfer energy from the left to the right. The FETs and diodes connected to the lower coil can transfer energy from the right to the left. Though the coils are shown as separate, they can be wound on a common core. Using this block in place of the bidirectional buck- boost converter 100 of FIG. 1 1 or FIG. 12 delivers an amplifier or power converter that falls within the scope of the present invention, and that is well-represented by the general depiction of FIG. 9.
• FIG. 21 illustrates the fact that an amplifier or power converter according to the present invention may also include a power adaption block 210, feeding a power reservoir.
• the power reservoir is shown as a capacitor 212 between VP and 0V.
• the adaption block 210 may be a DC-DC converter that draws energy from the battery at a smoothed rate. This can sidestep the power bottleneck that exists at many battery outputs.
• the DC-DC converter may perform some voltage boosting, so that the currents between the power reservoir 212 and the power stage 92 are not as high as they would otherwise be. This can ease component sizing.
• the amount of voltage boosting may be a function of prevailing conditions such as the amplifier's overall gain setting, or knowledge of the load, or knowledge of the prevailing signal level or of the evolving signal envelope.
• the adaption block may be an AC-to-DC converter.
• FET field-effect transistor
• the figures show circuits using FETs, alternatives are of-course possible, including other types of transistor (BJT, IGBT, HEMT) and various types of thyristor (SCR, GTO, MCT).
• Transistor segmentation techniques may be used, e.g. to improve efficiency at low signal levels.
• Boost amplifiers and power converters may incorporate the ability to read signals back from the load. This may be used to determine information about the load, to monitor the state of the load, to manage equalization filtering, to improve frequency response flatness, or for other purposes.
• Amplifiers and power converters as described with reference to the Figure 9 to 21 may be used in a wide range of devices . Examples include cars, desktop computers, notebook computers, tablet computers, mobile phones, loudspeaker subassemblies, actuation subassemblies, integrated positioners, medical implants, ultrasonic measurement equipment, RF transmitters, tactile feedback devices, smart panels, smart structures, and suchlike. Applications include audio amplification, actuation for tactile feedback in advanced user interfaces, ultrasonic amplification, RF amplification, other amplification, and power inversion.

Landscapes

• Engineering & Computer Science (AREA)
• Power Engineering (AREA)
• Amplifiers (AREA)

Priority Applications (3)

Applications Claiming Priority (2)

Publications (1)

Family

ID=55697757

Family Applications (1)

Country Status (6)

Cited By (2)

Families Citing this family (6)

Citations (2)

Family Cites Families (25)

• 2016
  • 2016-02-16 GB GBGB1602724.5A patent/GB201602724D0/en not_active Ceased
• 2017
  • 2017-02-15 EP EP17706856.6A patent/EP3417542B8/en active Active
  • 2017-02-15 CN CN201780017615.7A patent/CN109196775B/zh active Active
  • 2017-02-15 WO PCT/GB2017/050390 patent/WO2017141025A1/en active Application Filing
  • 2017-02-15 US US15/999,312 patent/US10897233B2/en active Active
  • 2017-02-16 TW TW106105126A patent/TWI731933B/zh active

Patent Citations (2)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events